Multiple alignment of protein sequences with repeats and rearrangements

Multiple sequence alignments are the usual starting point for analyses of protein structure and evolution. For proteins with repeated, shuffled and missing domains, however, traditional multiple sequence alignment algorithms fail to provide an accurate view of homology between related proteins, beca...

Full description

Bibliographic Details
Main Authors:	Phuong, Tu Minh, Do, Chuong B., Edgar, Robert C., Batzoglou, Serafim
Format:	Online
Language:	English
Published:	Oxford University Press 2006
Online Access:	https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1635250/

id	pubmed-1635250
recordtype	oai_dc
spelling	pubmed-16352502006-12-26 Multiple alignment of protein sequences with repeats and rearrangements Phuong, Tu Minh Do, Chuong B. Edgar, Robert C. Batzoglou, Serafim Computational Biology Multiple sequence alignments are the usual starting point for analyses of protein structure and evolution. For proteins with repeated, shuffled and missing domains, however, traditional multiple sequence alignment algorithms fail to provide an accurate view of homology between related proteins, because they either assume that the input sequences are globally alignable or require locally alignable regions to appear in the same order in all sequences. In this paper, we present ProDA, a novel system for automated detection and alignment of homologous regions in collections of proteins with arbitrary domain architectures. Given an input set of unaligned sequences, ProDA identifies all homologous regions appearing in one or more sequences, and returns a collection of local multiple alignments for these regions. On a subset of the BAliBASE benchmarking suite containing curated alignments of proteins with complicated domain architectures, ProDA performs well in detecting conserved domain boundaries and clustering domain segments, achieving the highest accuracy to date for this task. We conclude that ProDA is a practical tool for automated alignment of protein sequences with repeats and rearrangements in their domain architecture. Oxford University Press 2006-11 2006-11-26 /pmc/articles/PMC1635250/ /pubmed/17068081 http://dx.doi.org/10.1093/nar/gkl511 Text en © 2006 The Author(s)
repository_type	Open Access Journal
institution_category	Foreign Institution
institution	US National Center for Biotechnology Information
building	NCBI PubMed
collection	Online Access
language	English
format	Online
author	Phuong, Tu Minh Do, Chuong B. Edgar, Robert C. Batzoglou, Serafim
spellingShingle	Phuong, Tu Minh Do, Chuong B. Edgar, Robert C. Batzoglou, Serafim Multiple alignment of protein sequences with repeats and rearrangements
author_facet	Phuong, Tu Minh Do, Chuong B. Edgar, Robert C. Batzoglou, Serafim
author_sort	Phuong, Tu Minh
title	Multiple alignment of protein sequences with repeats and rearrangements
title_short	Multiple alignment of protein sequences with repeats and rearrangements
title_full	Multiple alignment of protein sequences with repeats and rearrangements
title_fullStr	Multiple alignment of protein sequences with repeats and rearrangements
title_full_unstemmed	Multiple alignment of protein sequences with repeats and rearrangements
title_sort	multiple alignment of protein sequences with repeats and rearrangements
description	Multiple sequence alignments are the usual starting point for analyses of protein structure and evolution. For proteins with repeated, shuffled and missing domains, however, traditional multiple sequence alignment algorithms fail to provide an accurate view of homology between related proteins, because they either assume that the input sequences are globally alignable or require locally alignable regions to appear in the same order in all sequences. In this paper, we present ProDA, a novel system for automated detection and alignment of homologous regions in collections of proteins with arbitrary domain architectures. Given an input set of unaligned sequences, ProDA identifies all homologous regions appearing in one or more sequences, and returns a collection of local multiple alignments for these regions. On a subset of the BAliBASE benchmarking suite containing curated alignments of proteins with complicated domain architectures, ProDA performs well in detecting conserved domain boundaries and clustering domain segments, achieving the highest accuracy to date for this task. We conclude that ProDA is a practical tool for automated alignment of protein sequences with repeats and rearrangements in their domain architecture.
publisher	Oxford University Press
publishDate	2006
url	https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1635250/
_version_	1611390463412011008

Multiple alignment of protein sequences with repeats and rearrangements

Similar Items